---
tags:
- text-to-image
- lora
- diffusers
- template:diffusion-lora
widget:
- text: '3DXLP5, young woman in blue denim jacket with a hat, in the style of , industrial texture, shot on 70mm, aggressive quilting, backlight, androgynous, wimmelbilder . --ar 68:103 --v 5.1 --style raw'
  output:
    url: images/111.png
- text: '3DXLP5, a man with dark brown hair and blue eyes is holding a gun in his right hand, aiming it at the camera with a cigarette in his mouth. He is wearing a black watch with a silver band and a gold ring on his left hand. His shirt is white with blue stripes and a gray t-shirt underneath. The background is a solid green, adding a pop of color to the scene.'
  output:
    url: images/222.png
- text: '3DXLP5, A close-up shot of a mans face, taken from a low-angle perspective. The mans eyes are a piercing blue, and his eyebrows are a darker shade of brown. His hair is a mix of brown and black, and hes sporting a goatee. He is wearing a green button-down shirt, and the background is a grayish gray.'
  output:
    url: images/333.png
base_model: black-forest-labs/FLUX.1-dev
instance_prompt: 3DXLP5
license: creativeml-openrail-m
---
![strangerzonehf/Flux-3DXL-Partfile-0005](images/xy.png)
<Gallery />

## Model description for 3DXL Partfile 0005 

Image Processing Parameters 

| Parameter                 | Value  | Parameter                 | Value  |
|---------------------------|--------|---------------------------|--------|
| LR Scheduler              | constant | Noise Offset              | 0.03   |
| Optimizer                 | AdamW  | Multires Noise Discount   | 0.1    |
| Network Dim               | 64     | Multires Noise Iterations | 10     |
| Network Alpha             | 32     | Repeat & Steps           | 18 & 2000 |
| Epoch                     | 14   | Save Every N Epochs       | 1     |

    Labeling: florence2-en(natural language & English)
    
    Total Images Used for Training : 13

## Best Dimensions & Inference

| **Dimensions** | **Aspect Ratio** | **Recommendation**       |
|-----------------|------------------|---------------------------|
| 1280 x 832      | 3:2              | Best                     |
| 1024 x 1024     | 1:1              | Default                  |

### Inference Range

- **Recommended Inference Steps:** 30–35

## Setting Up
```python
import torch
from pipelines import DiffusionPipeline

base_model = "black-forest-labs/FLUX.1-dev"
pipe = DiffusionPipeline.from_pretrained(base_model, torch_dtype=torch.bfloat16)

lora_repo = "strangerzonehf/Flux-3DXL-Partfile-0005"
trigger_word = "3DXLP5"  
pipe.load_lora_weights(lora_repo)

device = torch.device("cuda")
pipe.to(device)
```

## Trigger words

You should use `3DXLP5` to trigger the image generation.

## Download model

Weights for this model are available in Safetensors format.

[Download](/strangerzonehf/Flux-3DXL-Partfile-0005/tree/main) them in the Files & versions tab.